oe TREE Sta 43 een Sn (ANST S20 omen IBLE ee 6-1974) TEEE Recommended Practice for Testing Insulation Resistance of Rotating Machinery Sponsor Electric Machinery Committee ofthe IEEE Power Engineering Society “Approved 12 October 2000 American National Standards Institute Realfliemed 30 March 2006 ‘Approved 6 March 2000 IEER-SA Standards Board Abstract: This document describes the recommended procedure for measuring insulation resistance of ‘armature and field windings in rotating machines rated 1 hp, 750 W or greater. It applies to synchronous ‘machines, induction machines, do machines, and synchronous condensers, Contained within this IEEE Std 118-1978 (Reaff 1992), IEEE Standard Test Code for Resistance Measurements. IEEE Std 432-1992 (Reaff 1998), IBEB Guide for Insulation Maintenance for Rotating Electric Machinery (5 hp to less than 10 000 bp). IEE Std 433-1974 (Reaff 1991), TERE Recommended Practice for Insulation Testing of Large AC Rotating Machinery with High Voltage at Very Low Frequency. IBEE Std 434-1973 (Reaff 1991), IEEE Guide for Functional Evaluation of insulation Systems for Large High-Voltage Machines. IEEE Std 492-1999 IEEE Guide for Operation and Maintenance of Hydro-Generators. TEBE Std 510-1983 (Reaff 1992), JEEE Recommended Practices for Safety in High-Voltage and Hligh- Power Testing. ‘ASTAt publications ate aable fiom the Arian Society fr Testing and Dsterias, 10 Barr Harbor Drive, West Conshohocken, 'PA19628-2958, USA (utplivssin.oy), 2c publications ae avaliable fons th Seles Depron ofthe International Elecotechniel Commision, Cate Postale 131, 3 de Vung CILIZTI Genbve 2 SwiteroudSuste (twee). TEC pusetins are also avaable inthe Uaied States flom the Sales Deparment Auerican Netonal Standard netite 1 Wes 42nd Sree, 5th Floor, Nav York,NY 10036, USA. peseatly under evision “EEE publications ee avaiable om the lant of Bleich and HletooesEagicers, 45 Hoes Lane, PO, Box 1331 Piseatway, [NF 684551381, USA Otpelmmestandatd 20077). Srreseny under evison 2 ‘Copyright® 2000 IEEE, Al rights reservedeee RESISTANCE OF ROTATING MACHINERY ‘sid 43-2000 3. Definitions For the purposes ofthis recommended practice, the following terms and definitions apply. The IBEE Dictio- nary of Electrical and Blectronies Terms should be referenced for terms wot defined in this clause 31 absorption (polarization) current (Z4): A current resulting from molecular polarizing and electron rift, which decays with time of voltage application at a decreasing rate from a comparatively high intial 1 to neatly zero, and depends on the rype and condition of the bonding material used in the insulation system. 3.2 conduction current (fg): A current that is constant in time, that passes through the bulk insulation from the grounded surface fo the high-voltage conduetor, and that depends on the type of bonding material used in the insulation system, 3.3 eleetroendosmosiseffeet: A phenomenon occasionally observed, more often on older windings, when, in the presence of moisture, different insulation resistance values may be obtained when the polarity of the tester leads are reversed. Typically for older wet windings, the insulation resistance for reverse polarity, ‘where the ground lead is connected to the winding and the negative voltage lead to ground, is much higher ‘than for normal polarity 34 insulation resistance (IR): The capability ofthe electrical insulation of a winding to resist direct eur rent. The quotient of applied direct voltage of negative polarity divided by current across machine insulation, ‘corrected to-40 °C, and taken ata specified time () from start of voltage application. The voltage applicati time is usually 1 min (FR) or 10 min (Ro), however, other values cau be used. Unit conventions: values of 1 through 10 are assumed to be in minutes, values of 15 and greater are assumed to be in seconds. 35 geometric capacitive current (Jc): A reversible current of comparatively high magnitude and short ‘duration, which decays exponentially with time of voltage application, and which depends on the internal resistance of the measuring instrument and the geometric capacitance of the winding. 36 polarization index (PLL): Variation in the value of insulation resistance with time. ‘The quotient of the insulation resistance atime (i) divided by the insulation resistance at time (,). If times f and ty are not specified, hey are assumed to be 10 min and 1 min, respectively. Unit conventions: values of 1 through 10 ate assumed to be in minutes, values of 15 and greater are assumed to be in seconds (e.g., Pgs refers to FRead IRs) 3.7 surface leakage enrrent ({,): A current that is constant with time, and which usually exists over the sur- fice of the end-turns of the stator winding or between exposed conductors and the rotor body in insulated rotor windings, The magnitude of the surface leakage current is dependent upon temperature and the amount of conductive material, ie. moisture or contamination on the surface of the insulation, 4, Safety considerations Insulation resistance testing involves the application of high direct voltages to machine windings. These windings have capacitive and inductive properties that can lead to hazards that may not be readily apparent. tis not possible to cover all safety aspects inthis recommended practice and test personnel should consult IEEE Std 510-1983;° ASTM F835-97e1; manufacturers’ instruction manuals, and union, company, ad gov- ‘emment regulations. Stnformation on efiraces an be found in Cause 2 Copyright © 2000 IEEE: Al ights reserved 3eee $6 49-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION ‘Before any testing is conducted, the winding insulation must be discharged. It is not safe to begin testing ‘before the discharge current is zero and there is no discernible retura voltage (less than approximately 20 V) after the ground is removed. After completion of the test, the winding should be discharged through a suit- able resistor, sized to limit the instantaneous current to 1 A. A minimum discharge time, which is equal to four times the voltage application duration, is recommended. This time interval is based on the (resistive), (inductive), C (geometric capacitive), and absorptive characteristics ofthe circuit during charging (time of the application of the voltage) and discharging (elapsed time since the removal of the voltage source and subsequent grounding ofthe winding under test). Iris important to remember that the testing is not complete until the winding is discharged and there is no discemible voltage. It is recommended that subsequent ac hhigh-potential testing not be conducted until the winding Daring the test period, all appropriate safety measures for the voltages being used must be taken. For test ‘voltages 5000 V and above, the lead between the test set and the winding must be appropriately insulated ‘and spaced from ground; otherwise, surface leakage currents and corona loss may introduce errors in the test ‘data. For safety considerations, and to avoid measuring stray currents, the leads may be shielded. Restriction of personnel secess to the high voltages is mandatory. Use of personal protective equi recommended, as i¢ the use of hot sticks insulated adders, ete. IF accessible, the phase neutral and line ends of each winding should be connected together during the test fo minimize the effect of high-voltage surge reflections that may result from a winding failure. ‘The safety measures deseribed are by no means exclusive. These are meant only fo signify the nature of the hazards involved, It is the responsibility of the users of the test equipment to completely ascertain the possible hazards involved in the testing, to protect personnel from harm, and to eliminate the risk of damage to the equipment Insulation resistance—general theory ‘The insulatioa resistance of a rorating machine winding is a function of the type and condition of the insulat- ing materials used, as well as their application technique. In general, the insulation resistance varies propor- tionately with the insulation thickness and inversely withthe conductor surface area, Interpretation of insulation resistance measurements of machine windings and the recommended minimum vvalues of polarization index and insulation resistance are described in Clause 11 and Clause 12, respectively 5.1 Components of the measured direct current By definition, the insulation resistance is the quotient of the applied direct voltage across the insulation by the total resultant current at a given time. The total resultant current (/;) isthe sum of four differ cent currents: surface leakage ([,), geometric eapacitanee (Ic), conductance (Ig), and absorption (J) ‘An equivalent circuit for the various currents in an insulation resistance testis shown in Figure 1. ‘The geomettic capacitance current (Ic) usually does not affect the measurements, because it disappears by the time the first reading is taken at 1 min. 4 ‘Copyright ® 2000 IEEE. Alrights reserved. [ ise sone sonnsieconcuebcoassiNsaelees RESISTANCE OF ROTATING MACHINERY ‘st 43-2000 Soure of ‘ise vole "Foval Figure 1—Equivalent circuit showing the four currents monitored during an insulation resistance test ‘The absorption current (/,) or polarization current decays at a decressing rate. The current vs. time relation ship is a power function, show in Equation (1), It may be plotted on a log-log graph asa straight line. I= KE w where 1, is absorption current, K isa fimetion ofthe particular insulation system and appliad test voltage, © istime of applied direct voltage, niga fumetion of the particular insulation system. ‘The absorption current has two components. The first component is due to the polarization of the materials because the organic molecules, such as epoxy, polyester, and asphalt tend to change orientation inthe presence ofa direct electic field. Since these molecules have to strain against the attractive forces of other molecules, it usually takes several minutes after application of the electric field for the ‘molecules to become reoriented, and, thus, for the current-supplied polarizing energy to be reduced to almost zero. A second component of the absorption current is due to the gradual drift of electrons snd ions through most organic materials. These electrons and ions drift until they become trapped at the mica surfaces commonly found ia rotatiag machine insulation systems. Usually, for clean and dry rotating ‘machine insulation, the insulation resistance between about 30 s and a few minutes is primarily determined by the absorption current. ‘Since the absorption current is a property of the insulation material and the winding temperature, a specific absorption current is neither good nor bed, In insulation systems manufactured since about 1970 (usually thermosetting polyester or epoxy bonded), the value of the exponeat n ofthe absorption current, [,= KF", is different from the older thermoplastic (asphalt or shellac bonded) materials. This does not imply that more modern insulation materials are better because the absorption current is lower and the resulting insulat resistance is higher For example, polyethylene has essentially no absorption current, yer because ofits ther ‘would be completely unsuitable for application in most rotating machines. ‘The conduction eurrent (Ig) in well-bonded polyester and epoxy-mica insulation systems is essentially zero ‘unless the insulation has become saturated with moisture, Older insulation systems, such as asphaltie-mica or shellac mica-folium may have a natural and higher conduction current due to the conductivity of the tapes that back the mica, Copyright ® 2000 IEEE. Al ights reserved. 5vee 'S-49-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION ‘The surface leakage current (f,) is constant over time. A high surface leakage current, i.e, low insulation resistance, is usually caused by moisture or some other type of partly conductive contamination present in the machine. 5.2 Characteristics of the measured direct current ‘Comparing the change in insulation resistance or total current with the duration of the test voltage applica tion may be useful in appraising the cleanliness and dryness of a winding. IF the windings are contaminated (oF wet, the total current ([,) will be approximately constant with time, since f, and/or fg (see Figure 2) will bbe much larger than the absorption current ([,). If tke windings are clean and dry, the total current (/7) will wormally decrease with time (see Figure 3), since the total current is dominated by the absorption (.<., polar ization) current (/). 100, 10 Relative Current : Absorption (14) {Conductance (ia) [Capacitance (1<) t oo. 5 10 Time of Voltage Application (minutes) Figure 2—Types of currents for an asphaltic-mica insulation 5.8 Insulation resistance readings “The measurement of insulation resistance constitutes a direct-voltage test and must be resticted to a value appropriate tothe voltage rating of the winding and the basic insulation condition. This is particularly impor- tant in the case of small, low-voltage machines, or wet windings. Ifthe test voltage is too high, the applied test vollage may over stress the insulation, leading to insulation failure Insulation resistance tests are usually conducted at constant direct voltages of 500-10 000 V having negative polarity. Negative polarity is preferred to accommodate the phenomenon of electroendosmosis. Guidelines for test volinges are presented in Table 1, Readings of insulation resistance are taken after the test direct volt- ‘age has been applied for 1 min. 6 ‘Copyright © 2000 IEEE. Al rights rosowed. stoionesnsossensreaneamecieteneaseiaas(eee RESISTANCE OF ROTATING MACHINERY ‘sid 43-2000 100 Conductance (Ig)~ 0 104 Relative Current Total (/7) Absorption (4) ‘Time of Voltage Application (minutes) Figure 3—Types of currents for an epoxy-mica insulation with a relatively low surface leakage current and no conductance current Table 1—Guidelines for de voltages to be applied during insulation resistance test Winding rated | Tuswlation resistance test voltage direct voltage (V) <1000 500 1000-2500 500-1000 2501-5000 1000-2500 5001-12 000 2500-5000 12 000 000-10 000, "Rated lineto-line voltage for three-phase se machines, Hine-to-ground voltage for singlephase machines, sad rated rect voltage for de machines or field windings. 5.4 Polarization index readings ‘The measured insulation resistance will usually rapidly increase when the voltage is frst applied, and thea ‘gradually approach a relatively constant value as time elapses (see Figure 4), The readings of a dry winding in good condition may continue to inerease for hours with a continuously applied constant test voltage. For ‘older types of insulation, a reasonably steady value is usually reached in 10-15 min, Modem types of film ‘coated wire, as well a8 epoxy-mica or polyester-mica insulated stator windings, may approach a constant value of insolation resistance in 4 min or less, Ifthe winding is wet or dirty, alow steady value will usually bbe reached 1 min or 2 min after the test voltage is applied, ‘The polarization index is nommally defined as the ratio of the 10 min resistance value ({Rye) to the 1 main resistance value (UR,), (See Annex A for the use of other values.) The polarization index is indicative of the slope of the characteristic curve (see Figure 4) and can be used to assess the insulation condition (see Copyright © 2000 IEEE. Al ights reserved 7eee ‘Ste 48-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION Clause 11 and Clause 12). To provide greater accuracy around the 1 min point and to allow the data to be plotted on log paper, i is also common to take readings at other intervals such as 15 s, 30 s, 45 5, 1 min, 41.5 min, 2:min, 3 min, 4-min, .., and 10 min, Relative Resistance 1 Eee 10 Time of Voltage Application (minutes) Figure 4—Typical insulation resistance measurements for three different machines, 5.5 Discharge current “After the applied direct voltage is removed, a suitable discharge circuit should be provided (see Clause 4). ‘The discharge custent manifess itself in two components: @) A capacitive discharge current component, which decays nearly instantaneously, depending upon. the discharge resistance. b) The absorption discharge current, which will decay from a high initial value to neatly zero with the same characteristics as the initial charging current but with the opposite polarity. This decay may ‘ake more than 30 min depending on the insulation type and machine size ofthe test specimen. 6, Factors affecting insulation resistance 6.1 Effect of surface condition ‘The surface leakage current (f,) is dependent upon foreign matter, such as oi and/or carbon dust on the winding surfaces outside the slot. The surface leakage current may be significantly higher on large turbine ‘generator rotors and de machines, which have relatively large exposed creepage surfaces. There may also be Copyright® 2000 IEEE. Allright reserve, : 3 | ; | | 3 ' | i i i | :‘eee RESISTANCE OF ROTATING MACHINERY st6-43-2000 ‘an nerease in the surface leakage current on machines where a stres-control coating has been applied to the ‘endwindings. Dust (or salts) on insulation surfaces, which are ordinasily nonconductive when dry, may become partially ‘conductive when exposed to moisture or oil, and, thus, can lower the insulation resistance. IF the insulation resistance or polarization index is reduced because of contemination, it can usually be restored to an aceept- able value by cleaning and drying, 6.2 Effect of moisture Regardless ofthe cleanliness of the winding surface, if the winding temperature is at of belovr the dew point of the ambient air, a film of moisture may form on the insulation surface, which can lower the resistance or polarization index. The effect is more pronounced if the surface is also contaminated, or if ‘racks in the insulation are present. Note thatthe effects of moisture contamination on a healthy winding should not preclude obtaining acceptable readi ‘Some types of older winding insulation systems are hygroscopic (easily absorb water) and moisture may be ‘deawa into the body of the insulation from the humid ambient air, This is particularly true for the older asphaltic-miea and shelloe mica-folium insulating materials, xs well as for some insulating strips used between usinsulated copper conductors in large turbine generator rotors. Absorbed moisture increases the conduction current (7c) and significantly lowers the insulation resistance. In-service machines are usually at a temperature above the dew point. When tests are to be conducted on a machine that has been in service, the tests should be made before the machine winding temperature drops Delow the dew point. Machines that ae out-of-service (without space heaters) are frequently tested wisen the winding temperature is below the dew point and may have significantly lower than expected insulation resistance and polarization index readings due to moisture contamination (see Clause 11). Icmay be necessary to dry out these machines to obtain acceptable readings prior to returning these machines to service or performing high-voltage testing, For appropriate dry-out procedures, consult with the equipment manufacturer, The history of the machine, visual inspections, and other test results may help in assessing the potential risk of retumning to service a ‘machine with low insulation resistance andlor polarization index readings due to moisture contamination, It ig recommended that a machine with low PJ. and TR, readings not be subjected to further high-voltage resting. 6.3 Effect of temperature 6.3.1 General theory “The insulation resistance value for a given system, at any given point in time, varies inversely, on an expo- neatial basis, with the winding temperature, There is a coutrast between the temperature dependence of resistivity in metals and non-metallic materials, especially in good insulators. In metals, where there are ‘numerous free electrons, higher temperature introduces greater thermal agitation, which reduces the mean free path of electron movement with a consequent reduction in electron mobility and an increase in resstiv- ity. However, in insulators, an increase in temperature supplies thermal energy, which frees additional charge carriers tnd reduces resistivity. This temperature variation affects all of the current components identified in 5.1 except for the geometric capacitive current, The insulation resistance value of a winding depends upon the winding temperature and the time elapsed since the application of the voltage. The thermal mass of the ‘machine being tested is generally so large that tke temperature differential of the winding between the 1 min ‘and 10 min readings of the insulation resistance is negligible, except for measurements curing a dryout at ‘ated current, In order to avoid the effects of temperature in trend analysis, subsequent tests should be con- ‘ducted when the winding is ear the same temperature asthe previous test. However, ifthe winding temper- Copyright © 2000 IEEE, Al ghts reserved 9ieee 6 43-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION ature cannot be controlled from one test time to auother, it ia recommended that all insulation test values be conected to a common base temperature of 40 °C using Equation (2). Though the conected value is an approximation, this permits a more meaningful comparison of insulation sesistance values obtained at differ- ent temperatures “The conection may be made by using Equation (2) @ ‘Re is insulation resistance (in megohms) comected to 40 °C, Kz is insulation resistance temperature coefficient at temperature P°C (from 6.3.2 oF 6:33), Rp is measured insulation resistance (in megohms) at temperature T°C, For winding temperatures below the dew point, it is difficult to predict the effect of moisture condensation fon the surface, therefore an attempt to correct to 40 °C for trend aualysis would introduce an unacceptable error. In such cases, itis recommended thatthe history of the machine tested under similar conditions be the predominant factor in determining suitability for etum to service. However, since moisture contamination normally lowers the insulation resistance and/or polarization index readings, its possible to correct to. 40°C for comparison against the acceptance criteria (see Clause 12). ‘There is no effective means for converting the insulation resistance measured under a specific humidity to the insulation resistance that would occur at a different humidity. 6.3.2 Field measurements for determining Ky ‘The recommended method of obtaining data for an insulation resistance versus winding temperature curve is bby making measurements at several winding temperatures, al above the dew point, and plotting the results ‘on a semilogarithmic scale. When a logarithmic scale is used forthe insulation resistance and a linear scale for the temperature, the test poiuts should approximate a straight line that can be extrapolated to obtain the corrected value at 40°C. 6.3.3 Approximating Ky [fthe temperature effects on the insulation system under testis unknown, an approximate value forthe tem- perature coefficient Ky may be obtained by using Figure 5 for resistance halving for each +10 °C increment [Note that this is only an approximation and should not be used to calculate insulation resistance at very large temperature differentials from 40 °C or significant errors could result .NOTE-—insulation resistance halving fora 10 °C increas in temperature is based oa testing of some of the insulation systems ofthe late 1950s snd my not be silly tra forall nsultion systems. More recent measurements have in tiled a correction Factor for insulation resistance balving in the range $-20°C. A variation jn the Ky factor ca lend to ‘gnifeent erors in R, magnified by the differential between the winding lemperatare and 40 °C Kz can also be approximated for insulation resistance halving for a 10 °C rise in winding temperature by application of Equation (3). 10 ‘Copyright ® 2000 IEEE. Al ights reserved. session iii aceseee RESISTANCE OF ROTATING MACHINERY ‘std 43-2000 ient, K, Temperature Coeff Insulation Resistance Winding Temperature, °C Figure 5— Approximate insulation resistance coefficient, Ky, for insulation halving for 10°C rise in temperature. For example, ifthe winding temperature attest time was 35 °C, and the ingulation was such thatthe resis- ‘tance hnlved for every 10°C, then the Kp for correction t0 40 °C would be derived in the following way: Kp = (05) ® or Kp = (05) = (0.5) 1 = o707 0.5) 6.3.4 Polarization index correction ‘When the polarization index is used withthe insulation resistance to determine the insulation condition, itis aot necessary to make a temperature correction to the PZ. Since the machine temperature does not change appreciably between the 1 min and 10 min readings, the effect of temperature on the polarization index is ‘usually small. However, when the intial winding temperature is high, a reduction in the temperature of the insulation system during the test time may result in a substantcl increase in the insulation resistance ‘between the 1 min and 10 min readings due to the temperature effect (see 6.3.1). The resulting polarization index may be uncheracterstically high, in which case a repent measurement at or below 40 °C is recom ‘mended as a check of the P/.As stated in 6.2, if either the 1 mia or 10 min measurements are taken when the winding temperatuceis below the dew point, the effects of moisture contamination must be considered dur- {ng interpretation. 6.4 Effect of test voltage magnitude Guidelines for teat voltages are presented in Table 1 (see 5.3). The value of insulation resistance may decrease somewhat with an increase in applied voltage; however, for insulation in good condition and in a thoroughly dry state, substantially the same insulation resistance will be obtained for any test voltage up to the peak value of the rated voltage. {A sigoificant decrease in insulation resistance with an increase in applied voltage may be an indication of insulation problems, These problems may be due to imperfections or fractures ofthe insulation, aggravated by the presence of dirt or moisture; or the problems may be due to the effects of dirt or moisture alone or ‘result from other deterioration phenomena. The change in resistance is more pronounced at voltages consi ‘erably above rated voltage (see IEEE Std 95-1977). Copyright 2000 IEEE. Al ights reserved u(eee 14-43-2000 GEE RECOMMENDED PRACTICE FOR TESTING INSULATION 6.5 Effect of existing charge on winding resistance measurements ‘The insulation resistance measurements will be in error if residual charges exist in the insulation. Therefore, prior to measuring the insulation resistance, windings must be completely discharged, Measure the dis- charge current atthe beginning of the test to assure that the winding is completely discharged. A residual charge will show as a reverse deflection of the insulation resistance meter after comiections are made but before the voltage is applied. Any reverse deflection should be negligible. ‘After cessation of application of high direct voltage, grounding of windings is important for safety as well as for accuracy of subsequent tests. The grounding time should be a minimum of four times the charge (see 5.5). 7. Conditions for measuring insulation resistance Record the ambient temperature, relative humidity, dew point, winding temperature, length of time out-of service, test voltage, and connection arrangement at the time the testis performed. It is also important to convert the measurement to a 40 °C basis for future comparisons. (For converting insulation resistance val- ves fo this temperature, see 6,3). {is not necessary thatthe machine be at standstill when insulation resistance tests are being made. Its often desirable to make insulation resistance measurements when the winding is subject to centrifugal forces sim. ilar to those occurring in service. In certain eases, itis practical to make periodic insulation resistance mea- surements while machines are rotating short circuited for drying. Whenever machines are not at standstill ring measurement of insulation resistance, precautions should be taken to avoid damage to equipment and injury to personnel. ‘To obtain insulation resistance measurements for 2 direcily water-cooled winding, the water should be removed and the intemal circuit thoroughly dried. In some eases where water-cooled windings are used, the winding manufacturer may have provided @ means of measuring the insulation resistance without need for the coolant water to be drained. In general, if the water is not removed, then the conductivity of the water should be greater than 0.25 ys/em, More information should be available in the winding manufacturer's manual, 8. Winding connections for insulation resistance tests It is recommended, when feasible, that each phase be isolated and tested separately. Separate testing allows ‘comparisons to be made between phases. When one phase i tested, the other two phases should be grounded tothe same ground as the stator core or rotor body, ‘When testing all phases simultaneously, only the insulation to ground is tested and no testis made of the phase-to-phase insulation. The phase-to-phase insulation is tested only when one phase is energized and the ‘other phases are grounded. ‘The connection leads, brush rigging, cables, switches, capacitors, surge arresters, voltage transformers, and ‘other external equipment may greatly influence the insulation resistance reading, It is recommended tbat measurements ofthe insulation resistance be made with al exterual equipment disconnected and grounded, Inall eases, a common ground should be used to avoid any undesirable effects on the test results due to stray losses in the ground circuit, 2 ‘Copyright© 2000 IEEE. Al ights reserves. [eee RESISTANCE OF ROTATING MACHINERY sid 43-2000 9. Methods of measuring insulation resistance 9.1 Direct measurement Direct measurement of insulation resistance may be made with the following instruments: 4) Directindieating megohmmeter wit self-contained hand or power-driven generator b) _Directindicating megohmmeter with self-contained battery ©) Ditectindicating megoimmeter self-contained rectifier, using a regulated line powered supply 4) Resistance bridge with self-contained galvanometer and batteries 9.2 Calculated measurement Insulation resistance may be calculated from readings of a voltmeter and microammeter using au extemal (well-regulated) direct-voltage supply. NOTE—Recommended value of voltage regulation (Ine) is less than or equal 0.196, Deviations ftom this may lead 10 fnmbiguous rests due to unpredictable losses from the charging curents associated wih flactintions in the apptied vol- ‘age (se Annex C). The voltmeterammeter method is a simple method for the determination of the insulation resistance by measurement of the voltage impressed across the insulation and the current through it. A source of constant direct voltage is required, and the voltmeter must be selected to suit the maximum and minimum voltages that may be used, The ammeter is usually a mullirange microammeter selected to measure the full range of ‘currents that may be encountered atthe voltages used. "The microammeter must be on the highest range or short circuited during the first few seconds of eharge 30 that it will not be damaged by the capacitive charging current and the inital absorption current. When the ‘microammeter is at test voltage, precautions should be taken to ensure the safety of the operator. ‘The resistance is calculated from Equa (4). Ry = Boy ® where iy tvenslatn esitancen megan, By _ isthe voltmeter reading in volts, jy, isthe ammeter reading in microamperes (2 seconds after application of the test voltage, 10, Precautions ‘A finite amount of time is required to bring the voltage impressed on the insulation ¢o the desired test value, Fal test voltage should be applied as rapidly as possible and helé constant throughout the fst. “Test instruments in which the test voltage is supplied by motor-operated generators, batteries, or rectifiers fare usually used for making tests of over 1 min duration Itis essential that the voltage of any test source be ‘constant (0 prevent Guctuaifon in the charging current (see Annex C). Stabilization of the supplied voltage may be requited. CCopyiigh © 2000 IEEE, Al fights reserva. 3Beee Sta-43-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION “Where protective resistors are used in test insirumeuts, their effect on the magnitude ofthe voltage applied to te insulation under test should be taken into account. The voltage drop in the resistors may be an apprecia- ble percentage of the instrument voltage when measuring a low insulation re 11. Interpretation of insulation resistance and polarization index test results ‘The insulation resistance and polarization ndex tests can be used for two purposes: 8) The insulation resistance test history of a given machine, measured at uniform conditions so fer as the controllable variables are concerned, is recognized as a useful way of trending some aspects of the insulation condition over years. b) Estimation of the suitability of « machine for the application of appropriate overvoltage tests or for operation may be based on a comparison of present and previous PI. and/or IR, values. 11.1 Monitoring insulation condition 1f the insulation resistance history of the machine is available, comparison of the present test result with pre= vious tests will support concems about the insulation condition. It is important, however, to compare tests under similar conditions, that is, winding temperature, voltage magnitude, voltage duration, and relative ‘humility (see Clause 6). For comparison of tests conducted at different winding temperatures, the results should be corrected to the same temperature (see 6:3). [A sharp decline inthe 1R, or PJ from the previous reading may indicate surface contamination, moisture, or severe instlation damage, such as cracks. When a low PJ, occurs at an elevated temperature (above 60 °C), a second measurement below 40 °C, but above the dew point, is recommended as a check on the real insula- tion condition (see 6.3). For tests conducted under similar conditions, a steady increase in the /Ry, ic., a decrease in the absorption ‘current with age may indicate decomposition of the bonding materials, especially when the insulation mate- rials ae of the thermoplastic (asphaltic-mica or shellac mica-folium) type. 411.2 Suitability for operation or continued testing ‘When the insuletion resistance history is not available, recommended minimum values of the PL. or IR may bbe used to estimate the suitability ofthe winding for application of an overvoltage test or for operation (see Clause 12). If the IR, or PJ. is low because of dirt or excessive moisture, it may be improved to an accept able value by cleaning and drying. When drying insulation, the PJ. can be used to indicate when the drying process may be terminated, i, the PI. results have exceeded the recommended minimum. Ifthe JR, is low ‘due to severe insulation deterioration or damage, operation and overvoltage testing of the machine are not recommended. ‘Machines rated 10 000 kVA and less should have either a value of the polarization index or a value of the insulation resistance (at 40°C) above the minimum recommended values (see Clause 12) for operation or further overvoltage tests Machines rated above 10 000 kVA should have both the polarization index and the insulation resistance above the minimum recommended values (see Clause 12) for operation or further overvoltage testing. Ifthe 1R; value (at 40 °C) is greater than 5000 MO, the PJ. may be ambiguous and can be disregarded (see 1222), 4 Copyright© 2000 IEEE. Al ights reserve. ition aay\ece RESISTANCE OF ROTATING MACHINERY ia 43-2000 For varnished cambric, shellac mica-folium, or asphaltic stator windings, a very high PL, (for example, greater than 8) may indicate thatthe insulation has been thermally aged, and may have a high risk of failure. If physical inspection (tapping on the insulation, for instance) confirms that the insulation is dry and brittle, itis best not to attempt cleaning or overvoltage testing the winding. Failure may occur at any time if the ‘machine is returned to service. It may be possible o operate machines with PI. and IR, values lower than the recommended minimum val- ‘ues; however, it ie not recommended by this standard, Ia all cases where the test values fall below the recom- ‘mended minimam values, investigations should be undertaken to determine the cause of such Low readings, History of the winding, visual inspections, and other test results should be used to determine advisability of retuming the unit to service, 11.3 Limitations of the insulation resistance test Insulation resistance test data is useful in evaluating the presence of some insulation problems such as con tamination, absorbed moisture, or severe cracking; however, some limitations areas follows: 8) Insulation resistance of a winding isnot directly related to its dielectiie strength. Unless the defect is ‘concentrated, iis impossible to specify the valve of insulation resistance at which the insulation sys tem of a winding will fll b) Windings having an extremely Inrge end arm surface area, large or slow-speed machines, or machines with commutators may have insulation resistance values that are less than the recommended value, In these cases, historical trending of /R, is invaluable in evaluating insulation condition ©) A single insulation resistance measurement at one particular voltage does not indicate whether forcign matter is concentrated or distributed throughout the winding, 6) Direct-voltage measurements, such as the IR and PI. tests, may not detect internal insulation voids ‘caused by improper impregnation, thermal deterioration, or thermal cycling in form-wound stator coils (see Annex B). ©) Because insulation resistance tests are conducted while a machine is at standstill these tests will not detect problems due tozotation, such as loose cols, or vibration leading to endwinding movement. 12. Recommended minimum value of polarization index and insulation resistance 12.4 Minimum values ‘The recommended minimum PJ. and the recommended minimum value of /R of an ac or de rotating machine winding are the lowest values at which a winding is recommended for an overvolinge test or for operation In some cases, special insulating material or designs maay provide lower values. Minimum values for these desigas should be based on comparison with the historic test values. 12.2 Polarization index ‘The recommended minimum values of Pd, for se and de rotating machines are listed in Table 2. Table 2 is ‘based on the thermal class of the insulating materials and, with the exception of noninsulated field windings, applies to all insulating materials regardless of application, Copyright © 2000 IEEE lights reserved 1seee Std 43-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION Table 2—Recommended minimum values of polarization index for all machine components" insulation classes per IEC 60085-01: 1984 Thecmndar ng ancien |e ‘Class B 20 fs CF panera) oT 2d The PL test is not applicable to woninsulated field windings (see 12.2.1). ‘NOTE—I"the 1 min insulation resistance is above $000 MO, the ealenlsted P/. may not be meaningful. In such cases, the 2 may be disregarded ae a measure of winding conltion (see 1222) 12.2.1 Applicability of polarization index on field windings ‘The purpose of the insulation resistance and polarization index tests isto determine whether or not an insula- tion system is suitable for operation or overvoltage testing. The windings of some induction machinery rotors are often not insulated from the rotor body; therefore, a polarization index cannot be performed on these induction machinery rotor windings or field windings. If however, the rotor winding is insulated from. the rotor body, as in wound induction rotors and salient pole machines, a polarization index testis applica ble. The field windings of many very large turbine generators are made with exposed copper that is not ccucapsulated in insulation. Though isolated from ground and other components via insulating strips, the nmense surface area of the noninsulated copper does not exhibit an absorption current (J,), in comparison to the leakage current (J;), when subjected t0 a direct voltage. The absence of the absorption current alters the IR characteristic curve (see Figure 4) such that there will be very litle change in the JR value from the 41 min to the 10 min reading, Therefore, the PL., which describes the slope of the IR curve, is not applicable to noninsulated feld windings, On the other hand, many other types of field windings do not have appreciable amounis of exposed condue- tors. These designs use conductors that ate fully encapsulated in insulation and have a characteristic absorp- tion current (J,). For these machines, the PJ. ean be a worthwhile test for assessing the condition of the insulation system. The recommended minimum, based on the thermal class rating of the field win lation, should be used as a reference. 12.2.2 Applicability of polarization index when IR; Is greater than 5000 MO ‘When the insulation resistance reading obtained after the voltage has been applied for 1 min (Jf) is higher than 5000 MQ, based on the magnitude of applied direct voltage, the total measured current (2) cam be in the submieroampere range (see Figure 1). At this level of required test instrument sensitivity, small changes in the supply voltage, ambient humidity, test connections, and other noz-related components can realy affect the total current measured during the 1-10 min interval required for a PI. Because of these phenomena, ‘when the [Ry is higher thon 5000 MO, the PJ. may or may not be an indication of the insulation condition and is therefore not recommenced as an assessment tool 142.8 Insulation resistance ‘The minimum insulation resistance after 1 min, [Ry jg, For overvoltage testing or operation of se and de rmachine stator windings and rotor windings can be determined from Table 3. 16 CCopytigtt © 2000 IEEE. Al ights reserved. iaeee [RESISTANCE OF ROTATING MACHINERY sts 43-2000 ‘The actual winding insulation resistance to be used for comparison with Ry yin isthe observed insulation resistance, corrected 10 40°C, obtained by applying a constant direct voltage tothe entire winding for 1 min, ‘The minimum insulation resistance of one phase of a three-phase armature winding tested with the other two phases grounded should be approximately twice that of the entire winding. If exch phase is tested separately and guard circuits are used on the two phases not under test, the observed minimum resistance should be thee times the entire winding. Table 3—Recommended minimum Insulation resistance values at 40 °C (all values in M22) Minima ose ‘Test specimen For most windings made before about 1970, all itd windings, and coltece not described below ‘For most de armature and ao windiags but after about 1970 (form wound els) TR win” For most machines with rndom-wound stator coils and form- wound cols rated below 1 kV NOTES: 1—IR; piss the recommended minim insulation resistance, ia megohims, at 40 °C of the entre machine Winding ‘2-kF i the rated machine terminal to terminal voltage, in rms kV ‘Copytight © 2000 IEEE, Alrights reserves. 0eee Std 43-2000 IEEE RECOMMENDED PRACTICE FOR TESTING INSULATION Annex A (informative) Variants in polarization index ‘The polarization index (PI) is traditionally defined as the ratio of the 10 min insulation resistance (FRjg) to the 1 min insulation resistance (FR), tested ata relatively constant temperature. Ia older insulation materials, such a3 asphaltic-mica, the absorption currents often take 10 min or more to decay to nearly zero (see Figure 3). In more modem ingulation systems for form-wound stators, and especially in random-wound _machines, the absorption current may decay to nearly zero in 2-3 min (see Figure 4). Thus, for modem inst lation, some usets calculate a variant of the conventional PJ. The variants include, but are not limited to, those shown in Equation (A.1) and Equation (A.2).. PAL =IR\/TR oe (ay where PL is the polarization index, IR, is the insulation resistance reading after the application of voltege for t min, Ryo, is the insulation resistance reeding ater the application of voltage for 30 s. PL = IRS/IR, a2 where PI. isthe polarization index, ‘Rs, is the insulation resistance reading after the application of voltage for $ min, IR; isthe insulation resistance reading after the application of voltage for 1 min. “The distinguishing features are the shorter times the direct voltage is applied and thus the shorter time that the winding mast be grounded (see 6.5). Since in modem windings the absorption current is essentially zero after a few minutes, by using shorter times for the PI. rato, the test time can be considerably shortened with ‘out any loss of information about the degree of contamination or moisture absorption present, Another vari- lation is to record the insulation resistance every minule and discontinue the test when a stable (tree ‘consecutive readings) JK has been mensured, ‘There are limitations in applying these other ratios: 1) There is no standard for what ime use different ratios. tervals the IR values are to be ecorded. Different organizations b) ‘There is no agreed upon pass-fall riteria, as has been established forthe traditional PJ. ©) Users are encouraged to collect data employing shorter time ratios, to enable suitable pass-fil criteria to be developed in the future. 18 CCooyight © 2000 IEEE. At rights reserved, SN MMMM MAMMA eet(eee RESISTANCE OF ROTATING MACHINERY ‘ta «8-2000 Annex B (informative) Direct versus alternating voltage testing Direet-voltage testing is normally done by applying a direct-vollage source between the test specimen con- ductors and ground and using a de ammeter to measure the foal current. The ratio of the test voltage to the test current will rellect the total resistance between the test specimen and ground. Resistance is determined by Equation B.1). R= pl/A en © isesistivty of the material, E—islength ofthe path, A Iseross-sectional area, Because the resistivity values ofthe dirt, ol, and water that often contaminate the eudwinding areas of rotat- ing machinery are quite low, direct-voltage testing of a contaminated winding normally results in a high su face leakage current and subsequent low resistance reading, This property makes direct-voltage testing a viable method for determining the extent of contamination to an insulation system, In adation, ifthe insula- tion system uilizes « eotton-backed tape with mica as the primary electrical insulation, a direct-voltage test might reveal whether or not the cotton has absorbed moisture and has a lower resistivity. Note that most ‘windings manufactured after 1970 do not have these hygroscopic tapes, and a direct-voltage test will not ‘normally detect problems internal to the insulation system, such as thermal deterioration, Since the primary electrical insulation used in the design of Form-wound stator windings is mica, and mica hnag virtually infinite resistivity (us a good insulator), only one layer of mica tape would prohibit any direct ceurent, Therefore, if void exists within tbe insulation due to improper impregnation, thermal deterioration, or thermal cyeting, a direet-voliage test would be unable to detect it. If however, there exists a severe crack through the entire insulation, it ig possible that an electrical track would be established between the copper ‘conductors and ground, and would appear as a low resistance. ‘When a high alternating voltage is connected between the terminals of the test specimen and ground, the capacitance of the test specimen dominates the current. Capacitance is determined by Equation (B.2).. c= cara (82) where C — iscapacitance, is dieleotric permittivity of the material Ais cross-sectional area, is the thickness of the materiel Since the diclectic permittivity of an insulation system is greatly affected by the presence of voids and/or tvater, an alternating-voltage testis more sensitive than direct-voltage tests with regard to detection of inter- hal insulation problems associated with all types of insulation systems. Because of the different test capabil- ities, both a de and an ae test should be conducted to more completely assess the condition of an insulation system. Conyeght € 2000 IEEE. Al ights reserved. 19ese Sta 43-2000 REE RECOMMENDED PRACTICE FOR TESTING INSULATION Annex C (informative) Regulation in power supplies used for insulation resistance measurement ‘The regulation of power supplies used in insulation resistance (IR) measuring equipment is important ‘Decause variations inthe test voltage translate into variations in the measurement, This cau be seen by con- sidering the retatonship in Equation (C.1): iy ected en where iy is the eurent response, 4x() js variation in voltage, dt c is capacitance of the object under test ‘This variation in the curent due to voltage Nuctuations is translated into a variation in the insulation resis- tance (apparent resistance R,) according to Equation (C2): ae me 4° Ider itt) Ait where Ide the curent of terest de to the insulation resistance, it) isthe capctve caret, de sthe ppied direst vollege Combining these relationships gives the variation nthe apparent resianes shown in Equation (C3 t vac © Tex Cay ‘The regulation can be stated as a percentage of the applied voltage ancl calculated from values found in gen- eral practice. As most IR measurements are not more than 2% accurate, this accuracy can be used for the variation in the apparent resistance. For example, in stator winding insulation systems constructed using modern materials, the lowest IR value erally encountered is approximately 100 MO, atest voltage often used is 5 kV, and large machines bave a typical capacitance of approximately 0.25 uF. These numbers give am i(r) of 1 WA which, to maincain mea- surement accuracy, necessitates a dv/di << 4 Vis. The regulation would therefore be 4 V/s in S KV, or 0.08% regulated. 20 Copyright © 2000 IEEE. Aight reserves.RESISTANCE OF ROTATING MACHINERY. Iason ofthe supply, Le suscepbilty 50/60 Hz ripple in the supply may. tondwidth ofthe measurement limits ‘Most IR measurements are made at 0 these results apply 10 low frequency regue Somme ota os reeEE | IEEE Guide for Acceptance and | Maintenance of Insulating Oil in Equipment IEEE Power Engineering Society Sponsored by the IEEE Transformers Committee eee 3 Patk Avenue IEEE Std C57.106"-2006 |New York, NY 10016-5907, USA (Revision of IEEE Std C57.108-2002) ‘6 June 2007